Effect of Coke Breeze on Fissure Formation of Coke

Abstract

It is known that mixing the particle of inert material such as coke breeze to blending coal increases the mean size of lump coke. The size of lump coke is changed by the fissure formation, and it is assumed that the shape of lump coke is also changed by this mechanism. Therefore, the fissure formaiton of coke was investigated using the Gaudin-Meloy-Harris size distribution function. The coke blended with the coke breeze had larger grain size compared to the coke without mix of the coke breeze. In addition, the α value, which is the characteristic value of the Gaudin-Meloy-Harris size distribution function, reduced, suggesting that the shape of lump coke after breaking may be distorted. Therefore, as a result of analyzing coke fissures, it became clear that a change occurred in the fissure formation of the coke. The simulation results showed that the voidage of the coke packed layer was increased by increasing the coke grain size or distorting the shape of lump coke. It suggests that the coke produced from coal with coke breeze may increase the voids in the blast furnace.

1. Introduction

Coke plays a key role in ensuring air permeability in blast furnace and maintaining stable operation. Therefore, it is expected to produce a coke having a high strength and a large grain size.¹⁾ Among them, it has been reported that the intensity of the coke is maintained or increased by controlling the coal properties and the carbonization conditions.^2,3) On the other hand, in addition to controlling the coal properties⁴⁾ and the carbonization conditions,^5,6,7) it is known that the grain size of the coke is increased by blending an inert material such as coke breeze to blending coal. It is said that fissure formation in coke are changed and the grain size of coke is increased by blending coke breeze. And it is expected to affect the shape of lump coke as well by fissure formation changing. In this repot, it was evaluated how the fissure formation of coke was changed by blending coke breeze. In addition, we report the results of investigating fissure formation using the Gaudin-Meloy-Harris size distribution function.⁸⁾

2. Experiment

2.1. Testing Conditions

Table 1 shows properties of samples. Several kinds of coke prepared from coal blended with coke breeze were prepared using test ovens and evaluated. The carbonization was carried out under the following two conditions. For the test, a coke breeze having two particle sizes of 0.25 mm or less and 1 mm or less was used.

Table 1. Properties of samples.

(a) Can test
	breeze ratios [%]	breeze size [mm]	ASH [%]	VM [%]	logMF [logddpm]	Ro [%]	TI [%]
baseA	0	< 1.0	8.7	27.7	2.61	1.01	21.5
breezeA	1	< 1.0	8.7	27.3	2.59	1.00	22.3
baseB	0	< 1.0	8.6	27.7	2.65	1.01	21.4
breezeB	1	< 1.0	8.7	27.3	2.63	1.01	22.3
baseC	0	< 1.0	9.0	27.7	2.71	1.02	24.4
breezeC	1	< 1.0	9.1	27.3	2.69	1.01	25.3
baseD	0	< 1.0	8.7	27.6	2.74	1.02	20.1
breezeD	1	< 1.0	8.8	27.6	2.72	1.02	21.1

(b) Large oven test
	breeze ratios [%]	breeze size [mm]	ASH [%]	VM [%]	logMF [logddpm]	Ro [%]	TI [%]
base1	0	–	8.9	27.9	2.67	1.02	22.3
breeze1	1	< 0.25	8.9	27.6	2.65	1.01	22.6
base2	0	–	8.6	28.0	2.73	1.00	22.7
breeze2	2	< 0.25	8.7	27.5	2.70	0.99	24.1
base3	0	–	9.0	27.5	2.69	1.04	22.8
breeze3	3	< 1.0	9.1	26.6	2.61	1.01	25.1

1. Can test (can size: 235 × 235 × 300 mm)

A can packed with coal was charged into an oven heated to 1050°C. After that, it maintains the temperature of coke core for 2 hours after reaching the constant temperature, followed by discharging and smothering extinguishment.

2. Large oven test (1170 mm × 800 mm × 410 mm)

The retort filled with coal was charged into an oven and heated to 1050°C. After that, the core temperature has been reached at 950°C and maintained for 1.5 hours, followed by discharging and springing extinguishment.

2.2. Evaluation of Coke Properties

Measurements were performed according to JIS K 2151 (2004: Cokes-Test Methods).

2.3. Evaluation of Fissures

The surface of the coke cake in the test oven perpendicular to the oven wall was photographed and used for fissure analysis (see Fig. 1). Image analysis software WinROOF was used for image analysis. Fissure analysis was directed only to visible fissures. The fissures were classified as lateral fissures (in the direction of the oven height if the angle of the fissure was 45 degrees or more, and vertical fissure (in the direction of the oven width) if the angle of the fissures was less than 45°. However, at the top of coke cake, heat transfer from the top as well as from the oven walls. Therefore, fissures on top of the coke cake were excluded from this analysis because fissures direction at the top of coke cake was different from other parts.

Fig. 1.

Diagram of coke fissures.

2.4. Gaudin-Meloy-Harris Size Distribution Function

The under-sieving integral fraction D(D_p) for any sieve diameter D_p is given by the following equation.   

$$D(D_{\text{p}})=1-{\left(1-{\left(\frac{D_{\text{p}}}{D_{\text{p}0}}\right)}^{\alpha }\right)}^{\gamma }$$

α, γ: Characteristics value

D_p0: Initial grain size

In the case of coke, since the value of α is unknown, the value of α was calculated by the following procedure. The α value was determined so that the correlation equation between the logarithmic value of the weight fraction (1−D(D_p)) on the sieve of each sieve of 50, 38, 25, and 15 mm and the logarithmic value of D_p/D_p0 was the highest among the straight lines passing through the origin 0. In this test, the value of the oven width was used as the initial grain diameter D_p0.

2.5. Evaluation of Micro Cracks

The body of lump coke mass was hollowed out into a cylinder with a diameter of 5 cm and resin-molded to produce cores for measurements. Image capturing was performed using an electric zooming stereomicroscope SteeREO DIScovery V12 (manufactured by Karl Zeiss Microscopy Co.) and an image analysis software WinROOF was used to analyze the image. The linearity was calculated from the following equation.   

$$\text{[Linearity]}=\pi M^2/4S$$

S: Defect area [mm²]

M: Absolute maximum length [mm]

2.6. Assessment of the Filling Properties of Coke

The filling property of the coke was evaluated on the assumption that the properties and shapes of the coke changed depending on the presence or absence of coke breeze. The results of the laboratory tests of coke (base2) containing no coke breeze and coke (breeze2) containing coke breeze were used for the grain size distributions. The shape of lump coke was assumed to be cubic in base2. For the α values, it is assumed that the breeze2 is slightly distorted. Therefore, breeze2 was analyzed using two types of cube and cuboid.

2.6.1. Analysis Software

EDEM (manufactured by DEM-Solutions): Discrete Element Method.^9,10)

2.6.2. Condition

1000 kg of coke was dropped from the top of the container and filled into the container under the conditions described below. In order to exclude the effects of container bottoms and walls, the coke layers only inside the container were used to calculate the voidage. The voidage of the cylindrical portion having a radius of 300 mm from the center of the container was calculated at a height of 1000 mm from a point 250 mm from the bottom of the container.

2.6.3. Size Distribution

A 25–75 mm coke was used for the simulations. The grain size distribution of each fraction of 25–50 mm, 50–75 mm, and 75–100 mm was used from the test results of the large oven test, and the percentage was used.

2.6.4. Shape of Coke

Set with or without coke breeze (see Table 2).

Table 2. Material properties and size distribution.

	base2	breeze2	breeze2
Shape of Lump coke	cube (ratio 1:1:1)	cube (ratio 1:1:1)	cuboid (ratio 1:1:1.2)
size distributio [%]
50–75 mm	39.9	51.2	51.2
38–50 mm	40.0	35.8	35.8
25–38 mm	20.1	12.9	12.9
Material properties
Particle density [kg/m3]	1020
Static friction coefficient [−]	0.68
Poisson’s ratio [−]	0.22
Young’s modulus [MPa]	3
Restitution coefficient [−]	0.1
Rolling friction coefficient [−]	0.03

2.6.5. Container Size

Cylindrical container having a diameter of 1000 mm and a height of 5000 mm.

2.6.6. Material Property

Refer to Table 2 (measured data and literature¹¹⁾).

(The rolling friction coefficient and the restitution coefficient were indirectly determined by calculating the angle of repose which greatly influences the coefficient. It is calculated by DEM, fitting the calculated value of the angle of repose to the laboratory experiment value by sequentially changing the coefficient.)

3. Results and Discussion

Figure 2 shows the evaluation results of changes in the properties of coke carbonized by blending coke breeze with various coal samples. Different samples have different effects, but the DI tends to decrease and the size of lump coke tends to increase when coke breeze is blended. In this study, we report the effect of coke breeze on fissure formation of coke.

Fig. 2.

Effect of blending coke breeze on coke properties.

3.1. Micro Cracks

First, the results of evaluating the micro crack of coke are shown. The coke was carbonized in the can test. Cores for measurement were prepared from the coke. Defects in the coke core were classified into micro cracks and other defects by linearity and absolute maximum length, and the micro crack ratio was calculated.   

$$\begin{array}{l}\begin{array}{l}[\mathrm{P}\mathrm{e}\mathrm{r}\mathrm{c}\mathrm{e}\mathrm{n}\mathrm{t}\mathrm{a}\mathrm{g}\mathrm{e}\mathrm{o}\mathrm{f}\mathrm{m}\mathrm{i}\mathrm{c}\mathrm{r}\mathrm{o}\mathrm{c}\mathrm{r}\mathrm{a}\mathrm{c}\mathrm{k}\mathrm{a}\mathrm{r}\mathrm{e}\mathrm{a}\mathrm{\hspace{0.17em} }\\ (<\mathrm{\hspace{0.17em} }0.5,\mathrm{\hspace{0.17em} }0.5-1.0,\mathrm{\hspace{0.17em} }1-10\mathrm{\hspace{0.17em} }\mathrm{m}\mathrm{m}),\mathrm{\hspace{0.17em} }\%]\end{array}\hfill \\ \begin{array}{l}=\mathrm{\hspace{0.17em} }\mathrm{A}\mathrm{r}\mathrm{e}\mathrm{a}\mathrm{o}\mathrm{f}\mathrm{\hspace{0.17em} }\mathrm{l}\mathrm{i}\mathrm{n}\mathrm{e}\mathrm{a}\mathrm{r}\mathrm{i}\mathrm{t}\mathrm{y}\mathrm{\hspace{0.17em} }\text{[}>\mathrm{\hspace{0.17em} }7.5\text{]}\mathrm{\hspace{0.17em} }\mathrm{a}\mathrm{n}\mathrm{d}\mathrm{\hspace{0.17em} }\mathrm{a}\mathrm{b}\mathrm{s}\mathrm{o}\mathrm{l}\mathrm{u}\mathrm{t}\mathrm{e}\mathrm{\hspace{0.17em} }\mathrm{m}\mathrm{a}\mathrm{x}\mathrm{i}\mathrm{m}\mathrm{u}\mathrm{m}\mathrm{\hspace{0.17em} }\mathrm{l}\mathrm{e}\mathrm{n}\mathrm{g}\mathrm{t}\mathrm{h}\mathrm{\hspace{0.17em} }\\ [<\mathrm{\hspace{0.17em} }0.5,\mathrm{\hspace{0.17em} }0.5-1.0\mathrm{\hspace{0.17em} }\mathrm{m}\mathrm{m},\mathrm{\hspace{0.17em} }1.0-10\mathrm{\hspace{0.17em} }\mathrm{m}\mathrm{m}]/\mathrm{T}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}\mathrm{\hspace{0.17em} }\mathrm{a}\mathrm{r}\mathrm{e}\mathrm{a}\mathrm{\hspace{0.17em} }\mathrm{o}\mathrm{f}\mathrm{\hspace{0.17em} }\mathrm{c}\mathrm{o}\mathrm{r}\mathrm{e}\end{array}\hfill \end{array}$$

Figure 3 shows the relationship between the blending ratio of coke breeze and the micro crack ratio.⁷⁾ The blend of coke breeze tends to increase the percentage of micro-cracks of < 0.5 mm, 0.5–1.0 mm or 1 mm or less. The coke breeze does not shrink during the coal carbonization process. Therefore, the difference in shrinkage between the coke breeze and the surrounding coal causes stress. It is said to be the origin of the micro crack. Since the coke breeze blended is 1 mm or less, it is assumed that the micro crack generated is 1 mm or less. As a result, it is considered that micro cracks of 1 mm or less increased when coke breeze was blended.

Fig. 3.

Effect of blending coke breeze on micro cracks.

Figure 4 shows the relationship between the percentage of micro cracks and the mean size of coke. Figure 4 shows that the average size of lump coke increases as the micro crack percentage increases. This tendency is remarkable in the case of fine micro cracks of 0.5 mm or less. Figure 5 shows the relationship between the percentage of micro cracks and the f-value. In the drum test, cracks reduce as the number of revolutions increases, and surface fracture mainly occurs. Therefore, the f value, which is an index of the surface fracture, is calculated using the following equation.¹²⁾   

$$f=\frac{ln(D{I}_{15}^{150}/100)-ln(D{I}_{15}^{30}/100)}{120}$$

Fig. 4.

Relationship between percentage of micro cracks and mean size of coke.

Fig. 5.

Relationship between percentage of micro cracks and f-value.

Figure 5 shows no relationship between the micro crack of 0.5 mm or less and the f value. However, the increase in micro cracks of 0.5–1.0 mm resulted in a large f-value and tended to increase in surface fracture. These results suggest that blending coke breeze into coal increases micro cracks of < 1 mm or less in the lump coke, that even smaller micro cracks contribute to the increase in coke lump size, and that the effect of the micro cracks on the lowering in coke strength is negligible.

3.2. Observation of Fissure

Next, the fissure formation of coke cake with or without the blending of coke breeze was evaluated. The coke was carbonized using a large scale oven. Figure 1 shows a diagram of coke fissure image. Fissures in the coke cake include a main fissure propagate towards the center of the oven as coking progresses and a short lateral fissure or vertical fissure bifurcated from the main fissure. Figure 6 shows the relationship between the blending ratio of coke breeze and the change in the number of fissures. It was plotted the change in the number of fissures due to the blend of the coke breeze. Figure 6(a) shows that the number of main fissure does not change with the blending ratio of the coke breeze. Therefore, it is assumed that the influence of the blending coke breeze on the main fissure is small. The main fissure is a large fissure that extends from the oven wall to the center of the oven due to the differential shrinkage in the coke layer. The stress caused by the difference in shrinkage rate is large. On the other hand, the stress relaxation effect of micro cracks caused by blending coke breeze is small. As a result, it is assumed that the blending of the coke breeze could not inhibit a large fissure such as a main fissure. And, it is confirmed from Figs. 6(b), 6(c) that the number of small vertical fissure and lateral fissure tends to decrease by the coke breeze blending.

Fig. 6.

Relationship between percentage of blending coke breeze and change in the number of fissures.

From the above, it was clarified that the fissure reduction effect by the blending coke breeze had no effect on the large fissures such as the main fissures, and was effective only on the small fissures. From the above, it was clarified that the fissure reduction effect by the blending coke breeze had no effect on the large fissures such as the main fissures, and was effective only on the small fissures.

3.3. Vertical Fissure Ratio

Figure 7 shows the relationship between the coke breeze content and the ratio of vertical fissure, and Fig. 8 shows the relationship between the total number of fissures and the ratio of vertical fissure. In this analysis, the main fissure was excluded from the analysis. This is because, as described above, the coke breeze blending does not affect the number of main fissures. The number of fissures other than the main fissure was used to calculate the ratio of vertical fissure.   

$$\begin{array}{l}[Ratio\mathrm{\hspace{0.17em} }of\mathrm{\hspace{0.17em} }vertical\mathrm{\hspace{0.17em} }fissure]=\\ [Number\mathrm{\hspace{0.17em} }of\mathrm{\hspace{0.17em} }vertical\mathrm{\hspace{0.17em} }fissures(excluding\mathrm{\hspace{0.17em} }main\mathrm{\hspace{0.17em} }fissure)]/\\ [Total\mathrm{\hspace{0.17em} }number\mathrm{\hspace{0.17em} }of\mathrm{\hspace{0.17em} }fissures]\end{array}$$

Fig. 7.

Relationship between percentage of blending coke breeze and ratio of vertical fissures.

Fig. 8.

Relationship between total number of fissures and ratio of vertical fissure.

Figure 7 shows that blending of coke breeze increases the ratio of vertical fissure. It can also be seen from Fig. 8 that the ratio of vertical fissure increases as the total number of fissures decreases. When coke breeze is blended, the shrinkage difference between the coke breeze and the coal causes micro cracking, and the thermal stress is relieved. Since the stress relaxation effect is small, it is active for a small fissure generated by a small stress. A small fissure such as a secondary fissure generated from a main fissure is easily inhibited. Therefore, it is considered that the percentage at which fissure formation is inhibited is larger in the lateral fissure having many small fissures compared with the vertical fissure.

As a result, it is considered that the ratio of vertical fissure increased as the total number of fissures decreased. The increase in the ratio of vertical fissure means that the coke cake breaks into fissures anisotropic. It is suggested that the shape of lump coke may be distorted by increasing the anisotropic of the fissuring of the lump coke.

3.4. Application of Gaudin-Meloy-Harris Size Distribution Function

In order to express the change of fissure in the vertical and lateral direction ratio in terms of the particle size distribution, the fissure formation of coke was evaluated by using the Gaudin-Meloy-Harris size distribution function.

First, the relationship between the fissuring formation and the α value, which is the characteristic value of the Gaudin-Meloy-Harris size distribution function, was investigated using various minerals with different breaking condition. Table 3 shows various properties of minerals.¹³⁾ The properties of minerals can be classified by the number of specific fracture-prone surfaces called “cleavage surfaces” in addition to appearance such as color and shape and physical properties such as hardness and specific gravity. Cleavage is a property of breaking between atoms in a crystal structure along a location where the bonding force is weak. Assuming an ideal cleavage, minerals break depends on the number of cleavage surfaces. For example, it is said that if the cleavage surfaces are in one direction, the mineral peels in a layered form, and if the cleavage surfaces are in four directions, the mineral splits in an octahedral form.¹⁴⁾ That is, it is inferred that as the number of cleavage surfaces increases, the breaking condition becomes homogeneous, and as the number decreases, the cracking becomes distorted. Therefore, four kinds of minerals having different cleavage surfaces were prepared, and the relationship between the α value and breaking formation was investigated.

Table 3. Properties of minerals.

Mineral name	Formula	Crystal system	Color	Cleavage
Fluorite	CaF2	cubic	iris	4
Calcite	CaCO3	trigonal	achroma	3
Graphite	C	hexagonal	gray	1
Biotite	K(Mg,Fe)3AlSi3O10OH)2	monocline	black	1

Figure 9 shows the grain size distribution after crushing the mineral by 300 revolutions in a type I tester. The grain size distribution after crush differed greatly depending on the type of minerals. The grain size distribution of biotite and graphite in one direction of cleavage surfaces hardly changed even if they were crushed. On the other hand, in calcite and fluorite having a large number of cleavage surfaces, the large particle diameter side decreases, and the grain size distribution is broad. That is, it was confirmed that as the number of cleavage surfaces increases, the minerals tend to break and the lumps tend to become smaller. This grain size distribution was used to calculate the α value. The initial grain diameter was 100 mm, and the grain size distribution of the fractured minerals was calculated using sieves of 15, 9.5, and 5.6 mm.

Fig. 9.

Grain size distribution of minerals.

Figure 10 shows the relationship between the number of cleavage surfaces and the α value. It can be seen that the α value increases as the number of cleavage surfaces increases. As described above, in the case of minerals, the breaking becomes homogeneous as the number of cleavage surfaces increases. Therefore, it has been clarified that the α value increases as the cleavage surfaces increases, that is, the breaking condition becomes homogeneous. That is, the α value is an metric indicating the anisotropic of the breaking condition, and it is suggested that the breaking condition can be grasped by the α value. From the above, the breaking of the coke, that is the fissure formation of coke, was organized using this α value.

Fig. 10.

Relationship between number of cleavage surfaces and α value.

Figure 11 shows the relationship between the α value and the ratio of vertical fissure, which is the characteristic value of the Gaudin-Meloy-Harris size distribution function.

Fig. 11.

Relationship between α value and ratio of vertical fissure.

It can be seen that the ratio of vertical fissure of coke is increased and the α-value is decreased by blending coke breeze. As described above, the α value is considered to be an metric indicating the anisotropic of the breaking condition. That is, it is considered that the anisotropic of the breaking condition increases as the α-value decreases, and the shapes of lump coke after breaking become distorted. Also in this test result, the ratio of vertical fissure of coke became large when coke breeze was blended, and as a result, the anisotropic of fissure formation also became large, and the α value became small.

From the above, it was confirmed that the fissure formation of coke can be grasped by the α value which is the characteristic value of the Gaudin-Meloy-Harris size distribution function.

3.5. Simulation of Voidage Using DEM

It is considered that when the fissure formation of coke is changed, the coke shape is also changed. In such cases, it is assumed that the coke filling property (voidage at the time of coke filling) also changes. That is, the coke voidage is increased by both the effect of increasing the grain size of coke and the effect of distorting the shape of lump coke. From the above, it is expected that by blending coke breeze, the voidage increases and the air permeability in the blast furnace increases. Therefore, DEM was used to evaluate the effect of coke size and shape on the coke filling property. Evaluations were made by calculating the voidage when coke was filled in a cylindrical container having a diameter of 1000 mm, and comparing the each coke. Figure 12 shows the results.

Fig. 12.

Voidage of each coke filled.

It can be seen that the voidage of the breeze2 blending coke breeze is larger than that of the base2. In particular, when the shape of the lump is assumed to be a cuboid, the voidage is greatly increased. That is, it was confirmed that the filling property of coke changed not only by the grain size but also by the shape. When the coke breeze is blended, the grain size of coke becomes large, and the lump coke becomes distorted. As a result, it is considered that the breeze2 is less likely to be packed in the container compared with the base2, and the void is kept. Since the increase in voids is expected to contribute to the improvement in air permeability in the blast furnace, it is essential to find out whether the shape of lump coke is useful as a management metric of coke in addition to the intensity and the grain size. And the results of this simulation are only one condition, so we intend to systematically evaluate these conditions by increasing them in the future.

4. Conclusion

As a result of evaluating the effects of coke on fissure formation by blending coke breeze, an increase in micro cracks of 0.5 mm or less in coke was confirmed. As a result, it was confirmed that the generation of relatively small fissures occurring in the coke cakes was inhibited. As a result of evaluating the fissure formation of coke using the Gaudin-Meloy-Harris size distribution function, it was confirmed that the anisotropic of the fissure formation was increased by blending coke breeze. Simulations also confirmed that the grain size improvement and the distorting of lump coke by blending coke breeze had the effect of increasing voidage by decreasing the coke packing property in the blast furnace.

References

• 1)   T.  Ariyama,  M.  Sato,  T.  Sato,  S.  Watakabe and  R.  Murai: Tetsu-to-Hagané, 92 (2006), 114 (in Japanese).
• 2)   T.  Miyazu,  Y.  Okuyama,  N.  Suzuki,  T.  Fukuyama and  T.  Mori: Nippon Kokan Tech. Rep., 67 (1975), 125 (in Japanese).
• 3)   K.  Nishioka,  S.  Yoshida and  M.  Hariki: Tetsu-to-Hagané, 70 (1984), 358 (in Japanese).
• 4)   T.  Arima,  T.  Nishi and  T.  Okuhara: CAMP-ISIJ, 4 (1991), 1098 (in Japanese).
• 5)   R.  Nakajima,  S.  Hasebe,  H.  Taketomi,  Y.  Yamate,  Y.  Kawaguchi,  N.  Suzuki and  S.  Mitani: CAMP-ISIJ, 4 (1991), 149 (in Japanese).
• 6)   Y.  Suzuki,  J.  Kobayashi and  M.  Yokomizo: Coke Circular, 41 (1992), 97 (in Japanese).
• 7)   K.  Fukada,  T.  Yamamoto,  I.  Shimoyama,  T.  Anyashiki,  H.  Fujimoto and  H.  Sumi: Tetsu-to-Hagané, 93 (2007), 438 (in Japanese).
• 8)   C. C.  Harris: Trans. Metall. Soc. AIME, 235 (1966), 143.
• 9)   P. A.  Cundall and  O. D. L.  Strack: Geotechnique, 29 (1979), 47.
• 10)   N.  Mitarai: Bussei Kenkyu, 91 (2009), 371 (in Japanese).
• 11)   K.  Terui,  Y.  Kashihara,  T.  Hirosawa and  T.  Nouchi: Tetsu-to-Hagané, 103 (2017), 86 (in Japanese).
• 12)   H.  Joh and  S.  Ida: Seitetsu Kenkyu, (1966), No. 254, 6331 (in Japanese).
• 13)  The Association for the Geological Collaboration in Japan (AGCJ): Geological Dictionary, Heibon-sha, Tokyo, (1970), 368 (in Japanese).
• 14)   H.  Yokota and  K.  Kinoshita: Seimitsu Kikai, 31 (1965), 828 (in Japanese).